Automated analysis of immunohistochemical images based on curve evolution approaches

Abstract

The HER2/neu oncogene is notable both for its role in the pathogenesis of breast cancer and its role as a target of treatment. Qualitative or quantitative protein evaluation has been achieved using immunohistochemistry (IHC) on frozen and archival tissues, a widely adopted technique due to the standardization of the internal procedural steps and its easy and low-cost applicability to any laboratory. The goal of the present study is to introduce an efficient tool for the automated detection of HER2 protein overexpression in tissues, providing accurate, instant, yet objective interpretation outcomes through a formalized procedure. The comparison of results with classifications by specialists who evaluated the same tissue samples dataset confirms the efficiency and prospect of the methodology.

Image Features Based on Characteristic Curves and Local Binary Patterns for Automated HER2 Scoring

“…Curve-based automated analysis of immunohistochemical images have been tried in the past with limited success [15]. In this section, we introduce a novel feature vector called a characteristic curve.…”